Allow migrate on protnone reference with MPOL_PREFERRED_MANY policy
Subsystems (4)
v4 cover Allow migrate on protnone reference with MPOL_PREFERRED_MANY policy
This patchset is to optimize the cross-socket memory access with MPOL_PREFERRED_MANY policy.
To test this patch we ran the following test on a 3 node system.
Node 0 - 2GB - Tier 1 Node 1 - 11GB - Tier 1 Node 6 - 10GB - Tier 2
Below changes are made to memcached to set the memory policy, It select Node0 and Node1 as preferred nodes.
#include <numaif.h> #include <numa.h>
unsigned long nodemask;
int ret;
nodemask = 0x03;
ret = set_mempolicy(MPOL_PREFERRED_MANY | MPOL_F_NUMA_BALANCING,
&nodemask, 10);
/* If MPOL_F_NUMA_BALANCING isn't supported,
* fall back to MPOL_PREFERRED_MANY */
if (ret < 0 && errno == EINVAL){
printf("set mem policy normal\n");
ret = set_mempolicy(MPOL_PREFERRED_MANY, &nodemask, 10);
}
if (ret < 0) {
perror("Failed to call set_mempolicy");
exit(-1);
}Test Procedure:
===============
1. Make sure memory tiering and demotion are enabled.
2. Start memcached.
# ./memcached -b 100000 -m 204800 -u root -c 1000000 -t 7
-d -s "/tmp/memcached.sock"3. Run memtier_benchmark to store 3200000 keys.
#./memtier_benchmark -S "/tmp/memcached.sock" --protocol=memcache_binary
--threads=1 --pipeline=1 --ratio=1:0 --key-pattern=S:S --key-minimum=1
--key-maximum=3200000 -n allkeys -c 1 -R -x 1 -d 10244. Start a memory eater on node 0 and 1. This will demote all memcached pages to node 6.
5. Make sure all the memcached pages got demoted to lower tier by reading /proc/<memcaced PID>/numa_maps.
# cat /proc/2771/numa_maps
---
default anon=1009 dirty=1009 active=0 N6=1009 kernelpagesize_kB=64
default anon=1009 dirty=1009 active=0 N6=1009 kernelpagesize_kB=64
---6. Kill memory eater.
7. Read the pgpromote_success counter.
8. Start reading the keys by running memtier_benchmark.
#./memtier_benchmark -S "/tmp/memcached.sock" --protocol=memcache_binary --pipeline=1 --distinct-client-seed --ratio=0:3 --key-pattern=R:R --key-minimum=1 --key-maximum=3200000 -n allkeys --threads=64 -c 1 -R -x 6
9. Read the pgpromote_success counter.
Test Results:
=============
Without Patch
------------------
1. pgpromote_success before test
Node 0: pgpromote_success 11 Node 1: pgpromote_success 140974
pgpromote_success after test
Node 0: pgpromote_success 11 Node 1: pgpromote_success 140974
2. Memtier-benchmark result.
AGGREGATED AVERAGE RESULTS (6 runs)
==================================================================
Type Ops/sec Hits/sec Misses/sec Avg. Latency p50 Latency
------------------------------------------------------------------
Sets 0.00 --- --- --- --- Gets 305792.03 305791.93 0.10 0.18949 0.16700 Waits 0.00 --- --- --- --- Totals 305792.03 305791.93 0.10 0.18949 0.16700
======================================
p99 Latency p99.9 Latency KB/sec
-------------------------------------
--- --- 0.00
0.44700 1.71100 11542.69
--- --- ---
0.44700 1.71100 11542.69
With Patch
---------------
1. pgpromote_success before test
Node 0: pgpromote_success 5 Node 1: pgpromote_success 89386
pgpromote_success after test
Node 0: pgpromote_success 57895 Node 1: pgpromote_success 141463
2. Memtier-benchmark result.
AGGREGATED AVERAGE RESULTS (6 runs)
====================================================================
Type Ops/sec Hits/sec Misses/sec Avg. Latency p50 Latency
--------------------------------------------------------------------
Sets 0.00 --- --- --- --- Gets 521942.24 521942.07 0.17 0.11459 0.10300 Waits 0.00 --- --- --- --- Totals 521942.24 521942.07 0.17 0.11459 0.10300
=======================================
p99 Latency p99.9 Latency KB/sec
---------------------------------------
--- --- 0.00
0.23100 0.31900 19701.68 --- --- --- 0.23100 0.31900 19701.68
Test Result Analysis:
=====================
1. With patch we could observe pages are getting promoted.
2. Memtier-benchmark results shows that, with the patch, performance has increased more than 50%.
Ops/sec without fix - 305792.03
Ops/sec with fix - 521942.24
Changes:
V4
- Added an example in the "PATCH 2/2" commit message as per the discussion from V3.
V3:
- Added "* @vmf: structure describing the fault" comment for mpol_misplaced() to fix the warning.
https://lore.kernel.org/oe-kbuild-all/[email protected]/
-https://lore.kernel.org/lkml/[email protected]/ v2:
- Rebased on latest upstream (v6.8-rc7)
- Used 'numa_node_id()' to get the current execution node ID, Added 'lockdep_assert_held' to make sure that the 'mpol_misplaced()' is called with ptl held.
- The migration condition has been updated; now, migration will only occur if the execution node is present in the policy nodemask.
-https://lore.kernel.org/lkml/[email protected]/
Donet Tom (2):
mm/mempolicy: Use numa_node_id() instead of cpu_to_node() mm/numa_balancing:Allow migrate on protnone reference with
MPOL_PREFERRED_MANY policy
include/linux/mempolicy.h | 5 +++-- mm/huge_memory.c | 2 +- mm/internal.h | 2 +- mm/memory.c | 8 +++++--- mm/mempolicy.c | 36 +++++++++++++++++++++++++++--------- 5 files changed, 37 insertions(+), 16 deletions(-)
Signature · 1 line
2.39.3
PATCH v4 1/2 mm/mempolicy: Use numa_node_id() instead of cpu_to_node()
Instead of using 'cpu_to_node()', we use 'numa_node_id()', which is quicker. smp_processor_id is guaranteed to be stable in the 'mpol_misplaced()' function because it is called with ptl held. lockdep_assert_held was added to ensure that.
No functional change in this patch.
--- include/linux/mempolicy.h | 5 +++-- mm/huge_memory.c | 2 +- mm/internal.h | 2 +- mm/memory.c | 8 +++++--- mm/mempolicy.c | 14 ++++++++++---- 5 files changed, 20 insertions(+), 11 deletions(-)
include/linux/mempolicy.h +3 -2
@@ -167,7 +167,8 @@ extern void mpol_to_str(char *buffer, int maxlen, struct mempolicy *pol);/* Check if a vma is migratable */ extern bool vma_migratable (struct vm_area_struct *vma); -int mpol_misplaced (struct folio *,struct vm_area_struct *,unsigned long ); +int mpol_misplaced (struct folio *folio,struct vm_fault *vmf, +unsigned long addr);extern void mpol_put_task_policy (struct task_struct *);static inline bool mpol_is_preferred_many (struct mempolicy *pol)
@@ -282,7 +283,7 @@ static inline int mpol_parse_str(char *str, struct mempolicy **mpol)#endif static inline int mpol_misplaced (struct folio *folio, -struct vm_area_struct *vma, +struct vm_fault *vmf,unsigned long address) {return -1 ;/* no node preference */
mm/huge_memory.c +1 -1
@@ -1754,7 +1754,7 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf)*/ if (node_is_toptier (nid)) last_cpupid =folio_last_cpupid (folio); - target_nid =numa_migrate_prep (folio, vma, haddr, nid, &flags); + target_nid =numa_migrate_prep (folio, vmf, haddr, nid, &flags);if (target_nid ==NUMA_NO_NODE ) {folio_put (folio); goto out_map;
mm/internal.h +1 -1
@@ -1088,7 +1088,7 @@ void vunmap_range_noflush(unsigned long start, unsigned long end);void __vunmap_range_noflush (unsigned long start,unsigned long end); -int numa_migrate_prep (struct folio *folio,struct vm_area_struct *vma, +int numa_migrate_prep (struct folio *folio,struct vm_fault *vmf,unsigned long addr,int page_nid,int *flags);void free_zone_device_page (struct page *page);
mm/memory.c +5 -3
@@ -5033,9 +5033,11 @@ static vm_fault_t do_fault(struct vm_fault *vmf)return ret; } -int numa_migrate_prep (struct folio *folio,struct vm_area_struct *vma, +int numa_migrate_prep (struct folio *folio,struct vm_fault *vmf,unsigned long addr,int page_nid,int *flags) { +struct vm_area_struct *vma = vmf->vma; +folio_get (folio);/* Record the current PID acceesing VMA */
@@ -5047,7 +5049,7 @@ int numa_migrate_prep(struct folio *folio, struct vm_area_struct *vma, *flags |=TNF_FAULT_LOCAL ; } -return mpol_misplaced (folio, vma, addr); +return mpol_misplaced (folio, vmf, addr); }static vm_fault_t do_numa_page (struct vm_fault *vmf)
@@ -5121,7 +5123,7 @@ static vm_fault_t do_numa_page(struct vm_fault *vmf) last_cpupid = (-1 &LAST_CPUPID_MASK ); else last_cpupid =folio_last_cpupid (folio); - target_nid =numa_migrate_prep (folio, vma, vmf->address, nid, &flags); + target_nid =numa_migrate_prep (folio, vmf, vmf->address, nid, &flags);if (target_nid ==NUMA_NO_NODE ) {folio_put (folio); goto out_map;
mm/mempolicy.c +10 -4
@@ -2718,7 +2718,7 @@ static void sp_free(struct sp_node *n)* mpol_misplaced - check whether current folio node is valid in policy * * @folio: folio to be checked -* @vma: vm area where folio mapped +* @vmf: structure describing the fault * @addr: virtual address in @vma for shared policy lookup and interleave policy * * Lookup current policy node id for vma,addr and "compare to" folio's
@@ -2728,18 +2728,24 @@ static void sp_free(struct sp_node *n)* Return: NUMA_NO_NODE if the page is in a node that is valid for this * policy, or a suitable node ID to allocate a replacement folio from. */ -int mpol_misplaced (struct folio *folio,struct vm_area_struct *vma, +int mpol_misplaced (struct folio *folio,struct vm_fault *vmf,unsigned long addr) {struct mempolicy *pol;pgoff_t ilx;struct zoneref *z;int curnid =folio_nid (folio); +struct vm_area_struct *vma = vmf->vma;int thiscpu =raw_smp_processor_id (); -int thisnid =cpu_to_node (thiscpu); +int thisnid =numa_node_id ();int polnid =NUMA_NO_NODE ;int ret =NUMA_NO_NODE ; +/* +* Make sure ptl is held so that we don't preempt and we +* have a stable smp processor id +*/ +lockdep_assert_held (vmf->ptl); pol =get_vma_policy (vma, addr,folio_order (folio), &ilx);if (!(pol->flags &MPOL_F_MOF )) goto out;
@@ -2781,7 +2787,7 @@ int mpol_misplaced(struct folio *folio, struct vm_area_struct *vma,if (node_isset (curnid, pol->nodes)) goto out; z =first_zones_zonelist ( -node_zonelist (numa_node_id (),GFP_HIGHUSER ), +node_zonelist (thisnid,GFP_HIGHUSER ),gfp_zone (GFP_HIGHUSER ), &pol->nodes); polnid =zone_to_nid (z->zone);
Signature · 1 line
2.39.3
PATCH v4 2/2 mm/numa_balancing:Allow migrate on protnone reference with MPOL_PREFERRED_MANY policy
In reply to Donet Tomcommit bda420b98505 ("numa balancing: migrate on fault among multiple bound nodes") added support for migrate on protnone reference with MPOL_BIND memory policy. This allowed numa fault migration when the executing node is part of the policy mask for MPOL_BIND. This patch extends migration support to MPOL_PREFERRED_MANY policy.
Currently, we cannot specify MPOL_PREFERRED_MANY with the mempolicy flag MPOL_F_NUMA_BALANCING. This causes issues when we want to use NUMA_BALANCING_MEMORY_TIERING. To effectively use the slow memory tier, the kernel should not allocate pages from the slower memory tier via allocation control zonelist fallback. Instead, we should move cold pages from the faster memory node via memory demotion. For a page allocation, kswapd is only woken up after we try to allocate pages from all nodes in the allocation zone list. This implies that, without using memory policies, we will end up allocating hot pages in the slower memory tier.
MPOL_PREFERRED_MANY was added by commit b27abaccf8e8 ("mm/mempolicy: add MPOL_PREFERRED_MANY for multiple preferred nodes") to allow better allocation control when we have memory tiers in the system. With MPOL_PREFERRED_MANY, the user can use a policy node mask consisting only of faster memory nodes. When we fail to allocate pages from the faster memory node, kswapd would be woken up, allowing demotion of cold pages to slower memory nodes.
With the current kernel, such usage of memory policies implies we can't do page promotion from a slower memory tier to a faster memory tier using numa fault. This patch fixes this issue.
For MPOL_PREFERRED_MANY, if the executing node is in the policy node mask, we allow numa migration to the executing nodes. If the executing node is not in the policy node mask, we do not allow numa migration.
Example:
On a 2-sockets system, NUMA node N0, N1 and N2 are in socket 0, N3 in socket 1. N0, N1 and N3 have fast memory and CPU, while N2 has slow memory and no CPU. For a workload, we may use MPOL_PREFERRED_MANY with nodemask N0 and N1 set because the workload runs on CPUs of socket 0 at most times. Then, even if the workload runs on CPUs of N3 occasionally, we will not try to migrate the workload pages from N2 to N3 because users may want to avoid cross-socket access as much as possible in the long term.
In below table, Process is the Process executing node and Curr Loc Pgs is the numa node where page present(folio node)
===========================================================
Process Policy Curr Loc Pgs Observation
-----------------------------------------------------------
N0 N0 N1 N1 Pages Migrated from N1 to N0 N0 N0 N1 N2 Pages Migrated from N2 to N0 N0 N0 N1 N3 Pages Migrated from N3 to N0
N3 N0 N1 N0 Pages NOT Migrated to N3 N3 N0 N1 N1 Pages NOT Migrated to N3 N3 N0 N1 N2 Pages NOT Migrated to N3
------------------------------------------------------------
--- mm/mempolicy.c | 22 +++++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-)
mm/mempolicy.c +17 -5
@@ -1504,9 +1504,10 @@ static inline int sanitize_mpol_flags(int *mode, unsigned short *flags)if ((*flags &MPOL_F_STATIC_NODES ) && (*flags &MPOL_F_RELATIVE_NODES ))return -EINVAL ;if (*flags &MPOL_F_NUMA_BALANCING ) { -if (*mode !=MPOL_BIND ) +if (*mode ==MPOL_BIND || *mode ==MPOL_PREFERRED_MANY ) + *flags |= (MPOL_F_MOF |MPOL_F_MORON ); +else return -EINVAL ; - *flags |= (MPOL_F_MOF |MPOL_F_MORON ); }return 0 ; }
@@ -2770,15 +2771,26 @@ int mpol_misplaced(struct folio *folio, struct vm_fault *vmf,break ;case MPOL_BIND : -/* Optimize placement among multiple nodes via NUMA balancing */ +case MPOL_PREFERRED_MANY : +/* +* Even though MPOL_PREFERRED_MANY can allocate pages outside +* policy nodemask we don't allow numa migration to nodes +* outside policy nodemask for now. This is done so that if we +* want demotion to slow memory to happen, before allocating +* from some DRAM node say 'x', we will end up using a +* MPOL_PREFERRED_MANY mask excluding node 'x'. In such scenario +* we should not promote to node 'x' from slow memory node. +*/ if (pol->flags &MPOL_F_MORON ) { +/* +* Optimize placement among multiple nodes +* via NUMA balancing +*/ if (node_isset (thisnid, pol->nodes))break ; goto out; } - fallthrough; -case MPOL_PREFERRED_MANY :/* * use current page if in policy nodemask, * else select nearest allowed node, if any.
Signature · 1 line
2.39.3
Andrew Morton
maintainerQuoted 3 lines · On Mon, 25 Mar 2024 09:24:12 -0500 Donet Tom <[email protected]> wrote:
V4
- Added an example in the "PATCH 2/2" commit message as per the discussion from V3.
Thanks, I updated the changelogs in place.
Quoted 161 lines · Donet Tom
This patchset is to optimize the cross-socket memory access with
MPOL_PREFERRED_MANYpolicy.To test this patch we ran the following test on a 3 node system.
Node 0 - 2GB - Tier 1 Node 1 - 11GB - Tier 1 Node 6 - 10GB - Tier 2Below changes are made to memcached to set the memory policy, It select Node0 and Node1 as preferred nodes.
#include <numaif.h> #include <numa.h>unsigned long nodemask; int ret; nodemask = 0x03; ret = set_mempolicy(MPOL_PREFERRED_MANY | MPOL_F_NUMA_BALANCING, &nodemask, 10); /* If MPOL_F_NUMA_BALANCING isn't supported, * fall back to MPOL_PREFERRED_MANY */ if (ret < 0 && errno == EINVAL){ printf("set mem policy normal\n"); ret = set_mempolicy(MPOL_PREFERRED_MANY, &nodemask, 10); } if (ret < 0) { perror("Failed to call set_mempolicy"); exit(-1); }Test Procedure:
===============
1. Make sure memory tiering and demotion are enabled.
2. Start memcached.# ./memcached -b 100000 -m 204800 -u root -c 1000000 -t 7 -d -s "/tmp/memcached.sock"3. Run
memtier_benchmarkto store 3200000 keys.#./memtier_benchmark -S "/tmp/memcached.sock" --protocol=memcache_binary --threads=1 --pipeline=1 --ratio=1:0 --key-pattern=S:S --key-minimum=1 --key-maximum=3200000 -n allkeys -c 1 -R -x 1 -d 10244. Start a memory eater on node 0 and 1. This will demote all memcached pages to node 6.
5. Make sure all the memcached pages got demoted to lower tier by reading /proc/<memcaced PID>/numa_maps.# cat /proc/2771/numa_maps --- default anon=1009 dirty=1009 active=0 N6=1009 kernelpagesize_kB=64 default anon=1009 dirty=1009 active=0 N6=1009 kernelpagesize_kB=64 ---6. Kill memory eater.
7. Read thepgpromote_successcounter.
8. Start reading the keys by runningmemtier_benchmark.#./memtier_benchmark -S "/tmp/memcached.sock" --protocol=memcache_binary --pipeline=1 --distinct-client-seed --ratio=0:3 --key-pattern=R:R --key-minimum=1 --key-maximum=3200000 -n allkeys --threads=64 -c 1 -R -x 69. Read the
pgpromote_successcounter.Test Results:
=============
Without Patch
------------------
1.pgpromote_successbefore testNode 0: pgpromote_success 11 Node 1: pgpromote_success 140974
pgpromote_successafter testNode 0: pgpromote_success 11 Node 1: pgpromote_success 1409742. Memtier-benchmark result.
AGGREGATED AVERAGE RESULTS (6 runs)
==================================================================
Type Ops/sec Hits/sec Misses/sec Avg. Latency p50 Latency
------------------------------------------------------------------Sets 0.00 --- --- --- --- Gets 305792.03 305791.93 0.10 0.18949 0.16700 Waits 0.00 --- --- --- --- Totals 305792.03 305791.93 0.10 0.18949 0.16700======================================
p99 Latency p99.9 Latency KB/sec
-------------------------------------
--- --- 0.00
0.44700 1.71100 11542.69
--- --- ---
0.44700 1.71100 11542.69With Patch
---------------
1.pgpromote_successbefore testNode 0: pgpromote_success 5 Node 1: pgpromote_success 89386
pgpromote_successafter testNode 0: pgpromote_success 57895 Node 1: pgpromote_success 1414632. Memtier-benchmark result.
AGGREGATED AVERAGE RESULTS (6 runs)
====================================================================
Type Ops/sec Hits/sec Misses/sec Avg. Latency p50 Latency
--------------------------------------------------------------------Sets 0.00 --- --- --- --- Gets 521942.24 521942.07 0.17 0.11459 0.10300 Waits 0.00 --- --- --- --- Totals 521942.24 521942.07 0.17 0.11459 0.10300=======================================
p99 Latency p99.9 Latency KB/sec
---------------------------------------
--- --- 0.000.23100 0.31900 19701.68 --- --- --- 0.23100 0.31900 19701.68Test Result Analysis:
=====================
1. With patch we could observe pages are getting promoted.
2. Memtier-benchmark results shows that, with the patch, performance has increased more than 50%.Ops/sec without fix - 305792.03
Ops/sec with fix - 521942.24Changes:
V4
- Added an example in the "PATCH 2/2" commit message as per the discussion from V3.
V3:
- Added "* @vmf: structure describing the fault" comment formpol_misplaced()to fix the warning.
https://lore.kernel.org/oe-kbuild-all/[email protected]/
-https://lore.kernel.org/lkml/[email protected]/ v2:
- Rebased on latest upstream (v6.8-rc7)
- Used 'numa_node_id()' to get the current execution node ID, Added 'lockdep_assert_held' to make sure that the 'mpol_misplaced()' is called with ptl held.
- The migration condition has been updated; now, migration will only occur if the execution node is present in the policy nodemask.(quoted diff) +0
-2-https://lore.kernel.org/lkml/[email protected]/ -v1: https://lore.kernel.org/linux-mm/9c3f7b743477560d1c5b12b8c111a584a2cc92ee.1708097962.git.donettom@linux.ibm.com/#tDonet Tom (2):
mm/mempolicy: Usenuma_node_id()instead ofcpu_to_node()mm/numa_balancing:Allow migrate on protnone reference with
MPOL_PREFERRED_MANYpolicy
Quoted 6 lines
include/linux/mempolicy.h | 5 +++-- mm/huge_memory.c | 2 +- mm/internal.h | 2 +- mm/memory.c | 8 +++++--- mm/mempolicy.c | 36 +++++++++++++++++++++++++++--------- 5 files changed, 37 insertions(+), 16 deletions(-)
LGTM, Thanks! Feel free to add
in the future version.
--
Best Regards,
Huang, Ying